Load a fitted model, inspect it, and explore state reductions — entirely in your browser.
Setting µ > 0 models a growing population: a higher growth rate lowers the system mean residence time and expected decay rate, but does not affect the labeling curve itself nor the mean age.
M[i, j] is the transfer rate from state j into state i. Diagonal entries must be ≤ 0, off-diagonal entries ≥ 0, and each row must sum to ≤ 0 — edit a cell to see it checked live. Click a state's row/column header to mark it for removal.
These play both roles in the model: the observed pool weights s that make up the measured signal (used by every plotted function), and the pool-size vector p of the strong-mass-balance condition sᵀ(M + µI) ≤ 0. When that condition fails, the residence-time and decay-rate views are greyed out — it is exactly what guarantees both of them stay non-negative over the whole domain.
Click a state to mark it for removal, then reduce the model.